Pleading, Discovery, And Proof Of Sherman Act Agreements: Harmonizing Twombly And Matsushita

Matsushita and Twombly have reshaped federal standards for summary judgment and pleading, especially in antitrust cases. In earlier years, the Supreme Court had warned that “summary procedures should be used sparingly in complex antitrust litigation where motive and intent play leading roles, the proof is largely in the hands of the alleged conspirators, and hostile witnesses thicken the plot.” By 1969, Kenneth Dam could write that the Court appeared to have decided “that it is per se erroneous to grant summary judgment against a plaintiff in any treble-damage action.” But the Court’s acceptance of the Chicago School critique of antitrust law, beginning in 1977, brought with it a greater apprehension of the costs and potential for abuse of the private antitrust remedy. The same concerns prompted the Court to rethink plaintiffs’ burdens of production on a motion for summary judgment, and of pleading on a motion to dismiss. In framing these more exacting standards, the Court relied on prevailing models and economic understanding to set boundaries on the expense of discovery and the risks of error in jury trials. In place of the earlier permissive approach, Matushita and Twombly have created a kind of staged decision-theoretic inquiry. Courts now must ask at the pleading stage and, if necessary, again after discovery, whether the plaintiff has made a sufficient showing to justify the expected direct and indirect costs of further inquiry. Rather than pass contentions of fact through the stages of litigation with minimal review, the Matsushita/Twombly framework increases the burdens on pretrial motions, professedly to avoid unjustified direct costs of discovery and indirect costs of “false inferences” of agreement.

These procedural innovations have had special influence on the many antitrust cases that turn on the same substantive issue that Matsushita and Twombly themselves did: whether rivals in an alleged price-fixing (or other per se illegal) conspiracy ever formed an agreement within the meaning of Section 1 of the Sherman Act. Section 1 requires the plaintiff to establish both an agreement and a restraint of trade. When an alleged (or admitted) agreement can be characterized as ancillary to a productive purpose, the issue of restraint of trade may require pleading and proof of an actual harm to competition under some version of the rule of reason. My concern in this article, however, is with cases alleging per se violations of Section 1, such as naked horizontal price fixing, market allocation, and boycotts. In these cases, liability turns on whether the defendants coordinated their conduct by an agreement. As I have shown elsewhere, courts distinguish agreement from lawful independent or interdependent conduct by the sorts of communication the rivals used to coordinate nakedly anticompetitive actions. In order to make the necessary showing of tacit or express agreement, plaintiffs must allege and then produce evidence that suggests that the rivals coordinated their future competitive choices on issues like price by private communications. Resolutions of defendants’ motions to dismiss or for summary judgment on this issue alone frequently end Section 1 cases, either directly by granting the motion, as in Matsushita and Twombly themselves, or indirectly by denying the motion and engendering a settlement.

Under the present framework, whether the plaintiffs’ showing on either motion sufficiently suggests agreement depends on the court’s estimate of the plausibility of the inferences the plaintiffs ask the court to draw. Matsushita required courts, on a motion for summary judgment, to look for “evidence ‘that tends to exclude the possibility’ that the alleged conspirators acted independently.” In doing so, courts must take account of defendants’ “plausible motive[s]”: “the absence of any plausible motive to engage in the conduct charged is highly relevant to whether a ‘genuine issue for trial’ exists” because, if defendants “had no rational economic motive to conspire, and if their conduct is consistent with other, equally plausible explanations, the conduct does not give rise to an inference of conspiracy.” Thus, seemingly, not only must the defendants’ motives be plausible, the explanation inferred from the defendants’ conduct must be more plausible than alternative innocent explanations. Twombly imported plausibility into the pleading standard, requiring the complaint to contain “enough factual matter (taken as true)” to provide “plausible grounds to infer an agreement,” given the defendants’ likely incentives in the alleged circumstances.

But should plausibility mean different things in evaluating the plaintiffs’ showings on the two motions? The practical difference between the burdens of pleading and of production is discovery. On a motion to dismiss, the plaintiff must allege enough “facts” suggesting agreement to justify the expected costs of discovery. After discovery, at the summary judgment stage, the plaintiff must produce sufficient evidence supporting its allegations of agreement to justify the risk of a mistaken finding of liability. For some courts, the difference essentially ends with this formal distinction—the required showings are the same in substance. For other courts, this formal difference implies a substantive difference in the meaning of plausibility: if the plaintiff has to find evidence of private communication among the defendants’ officers, then the pleading standard plaintiff must meet in order to reach discovery should be less onerous than the evidentiary standard the plaintiff must meet in order to reach a jury. In this article, I critically examine these standards.

In the next Part, I review how the standards developed in the two decisional contexts and the relationship of both to discovery, identifying especially the points at which Twombly left open room for discretion in the district court. In Part II, I examine two recent cases that illustrate the substantial differences in stringency that courts (and individual judges) have shown in applying the plausibility inquiry at the pleading stage. In Part III, I consider some possible consequences of the more lenient approach, focusing on two recent cases in which courts denied motions to dismiss before discovery and then later resolved motions for summary judgment. In Part IV, I consider how predismissal discovery might narrow the disparities in application of Twombly, particularly in antitrust cases, by mitigating problems of estimation in evaluating plausibility.

Continue reading the full text of this article in PDF format.